WO2018193674A1 - Image processing device, image-capturing device, image printing device, control method of image processing device, and image processing program - Google Patents

Abstract

The present invention corrects an image more suitably. A control unit (40) corrects an image by reducing a first area in the image and enlarging a second area outside the first area, and determines, according to the distance from the first area to a first side of the image, whether to cut the first side in the image.

Description

Image processing apparatus, imaging apparatus, image printing apparatus, image processing apparatus control method, and image processing program

One embodiment of the present invention relates to an image processing apparatus that performs correction for deforming an image.

A technique for changing the impression of a subject by enlarging or reducing a part of the image by image processing is known. For example, by correcting the face of a person in the image to be small, it is possible to correct the image to a small-faced impression, to correct the image of a thin image by correcting the entire person to be thin, and to correct the legs longer In this way, there is a technique for correcting an image having a good style. Among these, as a technique for correcting a face in an image to be small, Patent Document 1 describes a technique for correcting a face to be small by setting an area including a human face area as a deformation area and reducing the deformation area. ing. In the technique described in Patent Literature 1, when the set deformation area is set outside the image, the correction set outside the image is not performed, so that deterioration in image quality is suppressed.

Japanese Patent Publication “JP 2009-104672 A”

However, the technique described in Patent Document 1 may or may not be corrected depending on the position of the person's face in the image. Therefore, for example, when the user performs shooting using a camera equipped with a function for correcting a small face, the effect intended by the user may not be obtained. Further, for example, when a plurality of human faces are captured in one image, correction is performed on the face of the person located at the center of the image, and the face of the person located at the end of the image is corrected. There is a case where only a part of the faces of some people has a small impression without correction, and a face of a person without correction has a relatively large impression.

One aspect of the present invention has been made in view of the above points, and is an image processing apparatus, an imaging apparatus, an image printing apparatus, an image processing apparatus control method, and image processing that can correct an image more appropriately. The main purpose is to provide a program.

An image processing apparatus according to an aspect of the present invention includes a correction processing unit that corrects the image by reducing a first region in the image and enlarging a second region outside the first region,
When the reduction direction of the first region is from the first side of the image toward the inside of the first region, the correction processing unit is responsive to a distance from the first region to the first side. Then, it is determined whether or not to cut out the first side of the image.

An image processing apparatus according to another aspect of the present invention includes a correction processing unit that corrects the image by reducing a first area in the image and enlarging a second area outside the first area. And the correction processing unit sets the distance from the first region to the first side when the reduction direction of the first region is from the first side of the image toward the inside of the first region. Accordingly, it is determined whether or not to change the reduction direction of the first area so as not to go from the first side of the image to the inside of the first area.

According to one aspect of the present invention, an image can be corrected more appropriately.

1 is a functional block diagram illustrating a configuration example of an image printing apparatus according to Embodiment 1 of the present invention. It is a flowchart which shows an example of the flow of the image processing which concerns on Embodiment 1 of this invention. It is a figure for demonstrating an example (processing example 1) of the image processing which concerns on Embodiment 1 of this invention. It is a figure for demonstrating an example (reference example 1) of image processing. It is a figure for demonstrating an example (processing example 2) of the image processing which concerns on Embodiment 1 of this invention. It is a figure for demonstrating an example (processing example 3) of the image processing which concerns on Embodiment 1 of this invention. It is a figure for demonstrating an example (processing example 4) of the image processing which concerns on Embodiment 1 of this invention. It is a figure for demonstrating an example (processing example 5) of the image processing which concerns on Embodiment 1 of this invention. It is a functional block diagram which shows the modification of the imaging device which concerns on Embodiment 1 of this invention. It is a figure for demonstrating an example (processing example 6) of the image processing which concerns on Embodiment 2 of this invention. It is a figure for demonstrating an example (processing example 7) of the image processing which concerns on Embodiment 2 of this invention. It is a figure for demonstrating an example (processing example 8) of the image processing which concerns on Embodiment 2 of this invention. It is a figure for demonstrating an example (processing example 9) of the image processing which concerns on Embodiment 3 of this invention. It is a figure for demonstrating an example (processing example 10) of the image processing which concerns on Embodiment 4 of this invention. It is a functional block diagram which shows one structural example of the control part which concerns on Embodiment 5 of this invention. It is a flowchart which shows an example of the flow of the image processing which concerns on Embodiment 5 of this invention. It is a figure for demonstrating an example (processing example 11) of the image processing which concerns on Embodiment 5 of this invention. It is a figure for demonstrating an example (processing example 12) of the image processing which concerns on Embodiment 5 of this invention.

Embodiment 1
Hereinafter, an embodiment (Embodiment 1) of the present invention will be described in detail with reference to FIGS.

<Configuration of image printing device>
First, an example of the configuration of an image printing apparatus 1 according to an embodiment of the present invention will be described with reference to FIG. FIG. 1 is a functional block diagram showing the configuration of the image printing apparatus 1 of the present embodiment. As illustrated in FIG. 1, the image printing apparatus 1 includes an imaging unit 10, an operation unit 20, a display unit 30, a control unit (image processing device) 40, a printing unit 50, and a storage unit 60. Note that the imaging unit 10 and the control unit 40 function as an imaging device.

The imaging unit 10 images a subject, and transmits the captured image (an image including a target region including the subject) to the control unit 40 as an input image. The target area is an area to be subjected to the reduction and enlargement processing by the control unit 40.

The operation unit 20 receives user input, and is realized by, for example, a touch panel or a mouse. For example, when the operation unit 20 is a touch panel, an input image is displayed on the display unit 30 including the touch panel.

The display unit 30 displays various images. The display unit 30 displays, for example, an image captured by the imaging unit 10 or an output image generated by an image correction unit 45 described later.

The control unit 40 controls the image printing apparatus 1 in an integrated manner. The control unit 40 functions as an image processing device that performs image processing on an image (input image) captured by the imaging unit 10 and generates an output image after processing (after correction). A specific configuration of the control unit 40 will be described later.

The printing unit 50 prints an output image (image) generated by the processing of the control unit 40. The printing unit 50 may further print an image drawn by the user via the operation unit 20 on the output image.

The storage unit 60 stores, for example, various control programs executed by the control unit 40, and includes a nonvolatile storage device such as a hard disk or a flash memory. For example, an input image and an output image are stored in the storage unit 60. The storage unit 60 may store parameters and the like necessary for the processing of the control unit 40 such as image processing (correction processing) and subject detection processing.

Note that the image printing apparatus 1 does not necessarily include the control unit 40 that functions as an image processing apparatus. For example, an external device that can be communicably connected to the image printing apparatus 1 may have the image processing function of the control unit 40.

Further, the image printing apparatus 1 may not include the imaging unit including the imaging unit 10 and the control unit 40 that functions as an image processing device. In this case, the imaging apparatus may function as an external apparatus that can be communicably connected to the image printing apparatus 1 or may not have the function. Furthermore, the image printing apparatus 1 may not include the imaging unit 10. In this case, for example, the imaging unit 10 functions as an external device that can be connected to the image printing apparatus 1 so as to be communicable.

≪Configuration of control part≫
Next, a specific configuration of the control unit 40 will be described with reference to FIG. In order to execute the functions of the image processing apparatus, the control unit 40 includes a subject detection unit (target region detection unit) 41, a correction region setting unit (correction processing unit) 42, a correction intensity setting unit (correction processing unit) 43, A cutout setting unit (correction processing unit) 44 and an image correction unit 45 are provided. The correction area setting unit 42, the correction intensity setting unit 43, the cutout setting unit 44, and the image correction unit 45 are collectively referred to as a correction processing unit.

The subject detection unit 41 detects a subject (target region) to be corrected from the input image input to the control unit 40. The subject detected by the subject detection unit 41 includes (1) a person, (2) a person's face, (3) each organ of the face such as eyes, mouth or nose, or (4) a face contour. For example, in the case of face detection, the subject detection unit 41 can detect a subject using existing technology, such as using information on a skin color area detected from an input image.

Note that the detection of the subject by the subject detection unit 41 may be manual. In other words, the user may detect the subject from the input image. In this case, for example, the subject detection unit 41 (subject selection unit) displays an input image on the display unit 30 and detects (selects) a subject to be corrected designated by a user input via the operation unit 20. To do. When there are a plurality of subjects in the input image, the subject detection unit 41 selects a subject to be corrected based on a user input.

For example, when the operation unit 20 is a touch panel, the subject in the input image displayed on the display unit 30 is selected by touching the touch panel. When the operation unit 20 is a mouse, a subject is selected based on the mouse operation.

Further, in this specification, description will be made assuming that correction processing is performed on an image captured by the imaging unit 10 (that is, an image including a subject). Specifically, description will be made assuming that reduction and enlargement processing is performed on a subject included in an image. However, the present invention is not limited to this, and the image to be corrected may not be an image captured by the imaging unit 10. In this case, the subject detection unit 41 detects a target object to be corrected (in other words, a target region including the target object) included in the image. That is, the subject detection unit 41 functions as a target region detection unit that detects a subject included in the image or a target region including a target other than the subject (in other words, a target region included in the image).

The correction area setting unit 42 sets a correction area to be corrected in the input image based on information indicating the subject detected by the subject detection unit 41 (eg, information indicating the position and size of the subject). The correction area setting unit 42 sets an area including the subject included in the input image as a reduced area (first area) in the input image. The correction area setting unit 42 sets an area outside the set reduction area, specifically, an area adjacent to the reduction area as an enlarged area (second area) in the input image. In other words, the correction area setting unit 42 sets a reduction area and an enlargement area as correction areas. Details will be described later.

The correction intensity setting unit 43 sets an enlargement ratio and a reduction ratio when correcting an image. The correction intensity setting unit 43 preliminarily uses a magnification α (<1) for reducing the reduction area set in the input image and a magnification β (> 1) for enlargement of the enlargement area set in the input image. A set value may be used, or a value set according to an input image or a user operation may be used.

The cut setting unit 44 sets a cut region to be cut from the image to be finally output based on the position and size of the reduced region set by the correction region setting unit 42. Details will be described later.

The image correction unit 45 corrects the input image by reducing the reduction region, enlarging the enlargement region, and cutting the cut region, and outputs the corrected image to the printing unit 50 and / or the display unit 30 as an output image.

Note that the image correction unit 45 has a first magnification (<1) for reducing the reduced area set in the input image and a second magnification (> 1) for enlarging the enlarged area set in the input image. As such, a preset value may be used, or a value set according to an input image or a user operation may be used.

≪Process flow by the control part≫
Next, the flow of image processing in the control unit 40 of the image printing apparatus 1 according to the present embodiment will be described with reference to FIG. FIG. 2 is a flowchart illustrating an example of the flow of image processing in the control unit 40.

When the subject detection unit 41 acquires an image captured by the imaging unit 10 as an input image, the subject detection unit 41 detects a subject to be corrected from the input image (S201: target region detection step). The correction area setting unit 42 sets a reduction area and an enlargement area based on information indicated by the subject detected by the subject detection unit 41 (S202: correction processing step). Further, the correction strength setting unit 43 corrects the correction strength (the reduction magnification α in the reduction region and the enlargement magnification β in the enlargement region) based on the information indicating the reduction region and the enlargement region set by the correction region setting unit 42. Is set (S203: correction processing step). Further, the cut setting unit 44 sets a cut region based on the information indicating the reduction region and the enlargement region set by the correction region setting unit 42 and the information indicating the correction strength set by the correction strength setting unit 43 ( S204: Correction processing step). Then, the image correction unit 45 generates an output image by reducing the reduction area, enlarging the enlargement area, and cutting the cut area (S205: correction processing step). Then, the control unit 40 outputs the generated output image to the printing unit 50 or the display unit 30.

≪Details of correction process≫
Next, details of the correction processing by the control unit 40 will be described with reference to processing examples.

<Processing Example 1>
FIG. 3A shows an input image 300 used in Processing Example 1, and a human face (target region) 301 is shown in the input image 300. Here, an example will be described in which a person's face 301 is a subject to be corrected and the person's face 301 is thinned by being reduced in the horizontal direction.

FIG. 3B shows correction areas (reduction area and enlargement area) set in the input image 300 by the correction area setting unit 42 in the processing example 1. As illustrated in FIG. 3B, the correction area setting unit 42 sets an area including the face 301 as a reduced area 302. In the processing example 1, the width of the reduced area 302 is the same as the width of the human face 301, but the present embodiment is not limited to this. In the processing example 1, the correction area setting unit 42 sets the reduced area 302 so as to cross the input image 300 in the vertical direction. However, in the present embodiment, the shape of the reduced area is not limited to this ( (Refer to Modification 1).

Further, the correction area setting unit 42 sets an area outside the reduction area 302 as the enlargement area 303. In Processing Example 1, an enlarged region 303 is provided on both sides of the reduced region 302 so as to sandwich the reduced region 302 therebetween. In Processing Example 1, the width of each enlarged region 303 is ½ of the width of the reduced region 302, but the present embodiment is not limited to this.

In the processing example 1, the image correction unit 45 reduces the reduction area 302 in the horizontal direction and enlarges the expansion area 303 in the horizontal direction. More specifically, the coordinate in the direction indicated by the arrow in FIG. 3B is the x coordinate, and the distance from the center line 304 of the reduction area 302 to the boundary between the reduction area 302 and the enlargement area 303 is d1. If the width from the center line 304 of the region 302 to the end of the enlarged region 303 opposite to the reduced region 302 is d2, the point P = (l + c, y) is copied to the point P ′ = (αl + c, y) when (1) l ≦ d1, and (2) the point P ′ = (αd1 + β (1-d1) + c when d1 <l ≦ d2. , Y). Here, α is a positive constant set by the correction intensity setting unit 43 as a magnification (enlargement / reduction ratio) with respect to the reduction region, and satisfies α <1. On the other hand, β is a positive constant set by the correction intensity setting unit 43 as a magnification (enlargement / reduction ratio) with respect to the enlargement region, and is defined by β = (d2−αd1) / (d2−d1), and β> 1 Meet.

For example, the correspondence between the x coordinate in the input image 300 and the x coordinate in the output image is a graph as shown in FIG. Since the magnification α is set to be less than 1, the inclination of the x coordinate of the output image with respect to the x coordinate of the input image 300 in the reduced area 302 is also less than 1. Therefore, as shown in FIG. 3C, the x coordinate of the output image is smaller than the x coordinate of the input image 300. Further, the x-coordinate of the output image becomes smaller than the x-coordinate of the input image 300 as it approaches the enlarged region 303. This means that the reduction area 302 is reduced toward the center line 304 of the reduction area 302 and the degree of reduction (the amount of movement of the pixel) increases as the distance from the center line 304 increases.

On the other hand, since the magnification β is set to be greater than 1, the inclination of the x coordinate of the output image with respect to the x coordinate of the input image 300 is also greater than 1. This means that the enlargement area 303 enlarges toward the center line 304 of the reduction area 302 and the degree of enlargement (the amount of movement of the pixel) decreases as the distance from the center line 304 increases.

The x coordinate of the input image matches the x coordinate of the output image at the end of the enlargement region 303 opposite to the reduction region 302. Thus, the area outside the enlarged area 303 in the input image 300 is neither enlarged nor reduced. Therefore, it can be said that the image correction unit 45 can generate the output image so that the image quality does not deteriorate.

Note that the image correction unit 45 only needs to determine the correction amount for each point P with respect to the reduced area 302 according to the distance r from the center line 304 of the point P. As described above, each point P Is not necessarily proportional to the distance r of the point P from the correction center c.

<Reference Example 1>
Next, a case where a sufficient correction area cannot be set in the input image when the subject to be corrected is located at the end of the image, or when the size of the correction target is larger than the image will be described. FIG. 4A shows an input image 400 used in Processing Example 2, and a human face 401 is shown in the input image 400. In the first reference example, unlike the first processing example, the human face 401 is shown at a position on the right side of the input image 400.

FIG. 4B shows correction areas (reduction area and enlargement area) set in the input image 400 by the correction area setting unit 42 in Reference Example 1. The correction area setting unit 42 sets an area including the face 401 as the reduced area 402. In Reference Example 1, the width of the reduced area 402 is the same as the horizontal width of the human face 401.

The correction area setting unit 42 sets the enlarged area 403 on both sides of the reduced area 402 so as to sandwich the reduced area 402. In Reference Example 1, the width of the enlarged region 403 on the left side of the reduced region 402 is ½ of the width of the reduced region 302 as in Processing Example 1, but the right end of the input image 400 is set to the right end of the reduced region 402. Therefore, the width of the enlarged area 403 on the right side of the reduced area 402 is smaller than that of the processing example 1.

In the reference example 1, when the correction process is performed using the reduction ratio and the enlargement ratio set under the same conditions as in the processing example 1, the relationship between the x coordinate of the input image 400 and the x coordinate of the output image is as shown in FIG. The graph is as shown in (c). As shown in FIG. 4C, the x-coordinate of the input image 400 does not match the x-coordinate of the output image on the outermost side (= image end) of the right enlarged region 403, and the image end of the output image The x coordinate of the input image 400 corresponding to the x coordinate does not exist. The point P = (l + c, y) corresponding to the point P ′ = (αd1 + β (1-d1) + c, y) at the image end of the output image is outside the image of the input image 400. Therefore, the image end of the output image refers to the outside of the input image 400, and the pixel value of the reference destination is indefinite. As a result, the image quality of the output image is degraded. Further, in the above-described prior art, when a sufficient correction area cannot be set as in the first reference example, the correction itself is not performed, and thus the target correction effect cannot be obtained.

<Processing example 2>
On the other hand, even when the control unit 40 according to the present embodiment cannot set a sufficient correction area, the correction area is set in advance by appropriately cutting off the end side of the output image. It is possible to generate a suitable output image without image quality deterioration at the edge of the image while correcting with the intensity.

(A) of FIG. 5 shows the input image 500 used in the processing example 2 and the correction regions (the reduction region 502 and the enlargement region 503) set in the input image 500 by the correction region setting unit 42. In the input image 500, since the human face (target area) 501 is located at the right end, the correction area setting unit 42 sets the reduced area 502 so as to contact the right side of the input image 500. The correction area setting unit 42 also sets the enlarged area 503 only on the left side of the reduced area 502, and cannot set the enlarged area 503 on the right side of the reduced area 502.

FIG. 5B shows an output image 504 that is corrected so as to reduce the width of the face 505 with respect to the input image 500. The width of the corrected human face 505 (= the corrected width of the reduced area 502) is corrected to be narrower than the width of the corrected human face 501 (= the reduced width of the reduced area 502). . At this time, the cut setting unit 44 sets an area 508 indicated by shading at the right end of the output image 504 as a cut area. An area 508 is an area where the reference destination is outside the image of the input image 500, and the pixel value is indefinite. That is, the point P = (l + c, y) corresponding to P ′ = (αd1 + β (1-d1) + c, y) in the region 508 is outside the image of the input image 500. As a result, as shown in FIG. 5C, the image correction unit 45 can generate an output image 509 in which a region 508 in which the pixel value is indefinite is cut out from the output image 504. As a result, it is possible to obtain a suitable output image that does not include a region with an indefinite pixel value.

As described above, when the sufficient enlargement area can be set, the control unit 40 compensates for the reduction effect of the reduction area by enlarging the enlargement area as in Processing Example 1, and the image quality is deteriorated. An output image can be generated so as not to occur. That is, the area outside the enlarged area in the image can be an area that is neither reduced nor enlarged. If the control unit 40 cannot set a sufficiently large area, the control unit 40 sets a cut area and cuts the cut area as in Processing Example 2 so that an area with an indefinite pixel value is included. A suitable output image can be obtained.

In order to obtain such an effect, when the reduction direction of the reduction area is directed from the one side to the inside of the reduction area, the cut setting unit 44 determines whether the reduction direction of the reduction area depends on the distance from the reduction area to the first side. It is determined whether or not the first side of the image is to be cut out. When the first side is cut out, a cut area is set on the first side. In other words, the cropping setting unit 44 has a small space between the reduced area (first area) and one side (first side) of the input image (for example, the reduced area and one side of the input image are in contact with each other). If it is, the cut area is set so as to cut the first side of the image. Otherwise, the cut area is not set. The cutout region can be, for example, an end portion on the one side of the corrected image.

It should be noted that “cutting a certain side of an image” means removing an area having a certain width from a certain side included in the image from the image. The “reduction direction of the reduction area (first area)” means a direction in which pixels in the reduction area move when the reduction area is reduced. Further, “the reduction direction of the reduction area is directed from the one side to the inside of the reduction area” means that the pixel on the one side of the reduction area moves to the inside of the reduction area. At this time, if the reduction direction is represented by a vector and decomposed into a vector in a certain direction and a vector in a direction orthogonal to the certain direction, the obtained vector in the certain direction becomes a positive direction.

When the reduction direction of the reduction area is from a certain side toward the inside of the reduction area, the reduction area is reduced away from a certain side. Therefore, in order to compensate for the reduction effect, It is preferable to set a sufficient enlarged region on the side of the image. However, when the interval between the reduced area and the certain side is narrow (for example, when the reduced area is in contact with the certain side of the input image), it is not possible to set a sufficient enlarged area. In such a case, as described in Reference Example 1, a region with an indefinite pixel value is generated on the certain side of the reduced region. Therefore, the cropping setting unit 44 sets a cropping region so that the certain side of the image is cropped, thereby obtaining a suitable output image that does not include a region with an indefinite pixel value.

On the other hand, when the interval between the reduced area and the certain side is wide, a sufficient enlarged area can be set. Therefore, the cut setting unit 44 does not set the cut area so that the image quality does not deteriorate. An output image can be generated.

In other words, when the reduction direction of the reduction area (first area) is directed from a certain side to the inside of the reduction area, the cut setting unit 44 performs correction according to the distance from the reduction area to the certain side. It is determined whether or not to cut out the certain side in the subsequent image, and thus the image can be corrected more appropriately.

In one embodiment, when the reduction direction of the reduction area is directed from a certain side to the inside of the reduction area, the cut setting unit 44 sets the most from the center of the reduction area (first area) to the certain side in the reduction area. The distance from the reduced region to a certain side (distance from the region closest to the certain side in the reduced region to the certain side) is set to dx, and the scaling ratio of the reduced region is defined as d1. When α is set and the enlargement / reduction ratio of the enlargement region is β, when the following expression (1) is satisfied, a cropping area is set on the certain side of the image, and when expression (1) is not satisfied The cutting area is not set.

(1-α) d1> (β-1) dx (1)
More specifically, in one embodiment, the cropping setting unit 44 sets (1−α) d1− (β−1) dx to the certain side of the corrected image when Expression (1) is satisfied. Set the crop area for the width. Thereby, the control part 40 can cut out an appropriate area | region from the image after correction | amendment. Note that the width of the cut region set by the cut setting unit 44 may be smaller than (1-α) d1- (β-1) dx, and even in that case, the region where the pixel value is indefinite is reduced. A suitable output image can be obtained. Further, the width of the cut region set by the cut setting unit 44 may be wider than (1-α) d1- (β-1) dx, and even in this case, a region with an indefinite pixel value is included. A suitable output image can be obtained. However, it is preferable that the cut setting unit 44 is provided so that the cut area is outside the reduced area after reduction.

In the above-described processing example, the reduced area is set as a rectangular area that crosses the input image in the vertical direction, and the configuration in which the reduced area is reduced in the horizontal direction has been described. However, the present embodiment is not limited to this. The correction method is not particularly limited. For example, (A) a configuration in which reduction and enlargement is performed toward a specific straight line (configuration in the above-described processing examples 1 and 2), and (B) a reduction is performed toward a specific point. Various methods such as a configuration for enlarging (configurations of processing examples 3 to 5 to be described later) can be used. (A) Also in the configuration in which reduction and enlargement are performed toward a specific straight line, the straight line may be the center line of the reduction region, or may be another straight line, and the direction of the straight line may also be the vertical direction of the image. It may be a direction, a horizontal direction, or an oblique direction. Further, the shape of the reduced area and the enlarged area is not limited to a rectangle, and a part thereof may be configured by a curve. (B) Also in the configuration in which reduction and enlargement are performed toward a specific point, the point may be the center of the reduction region, or may be another point. May be isotropic, or may be isotropic. Further, the shape of the reduced area and the enlarged area is not limited to a circle, and may be an ellipse or a polygon. Hereinafter, as an example of a configuration that reduces and enlarges toward a specific point (B), a configuration that reduces and enlarges toward the center of a circular reduction region will be described with a processing example.

<Processing example 3>
FIG. 6A is a diagram illustrating an example of a reduced area 551 and an enlarged area 552 set by the correction area setting unit 42 with respect to the target area 550 that is a correction target. As shown in FIG. 6A, the reduced area 551 is a circular area having a radius d1 centered on the correction center c, and the enlarged area 552 is a circular area having a radius d2 centered on the correction center c. This is an area outside the reduced area 551. As described above, the shapes of the reduced region 551 and the enlarged region 552 are not limited to a circle.

The correction performed by the image correction unit 45 on the input image is (1) a reduction of the reduction area 551 toward the correction center c and (2) a correction of expanding the enlargement area 552 toward the correction center c. More specifically, a point P = (r cos θ, r sin θ) + (c1, c2) where the distance from the correction center c is r and the direction viewed from the correction center c = (c1, c2) is θ. (1) When r ≦ d1, the point P ′ = (r′cos θ, r′sin θ) + where the distance from the correction center c is r ′ = αr and the direction viewed from the correction center c is θ. (C1, c2), (2) When d1 <r ≦ d2, the distance from the correction center c is r ′ = βr− (β−α) d1, and the direction viewed from the correction center c is θ Is a correction copied to a point P ′ = (r′cos θ, r′sin θ) + (c1, c2). Here, α is a positive constant set by the correction intensity setting unit 43 as the enlargement / reduction ratio for the reduction region 551 and satisfies α <1. On the other hand, β is a positive constant set by the correction intensity setting unit 43 as the enlargement / reduction ratio for the enlargement region 552, is defined by β = (d2-αd1) / (d2-d1), and satisfies β> 1. .

For example, when α = 0.9 and β = 1.1 (d2 = 2d1), the distance r from the correction center c of the point P before correction and the correction center c of the point P after correction are The relationship with the distance r ′ is as shown in the graph of FIG. The enlarged region 552 is enlarged inward while maintaining the outer periphery, and the inner periphery of the corrected enlarged region 552 coincides with the outer periphery of the corrected reduced region 551. Further, the area outside the enlarged area 552 does not change before and after correction.

Note that the image correction unit 45 only needs to determine the correction amount for each point P in accordance with the distance r from the correction center c of the point P with respect to the reduced region 551. As described above, each point P Is not necessarily proportional to the distance r of the point P from the correction center c.

Further, as described above, the image correction unit 45 may be anisotropically reduced by changing the correction amount for each point P in accordance with the angle θ with respect to the reduction region 551. .

<Processing example 4>
FIG. 7A shows an input image 600 used in Processing Example 4, and a human face (target region) 601 is located at the right end. In Processing Example 4, the control unit 40 performs correction to reduce the face 601 of the person toward the center of the face. The correction area setting unit 42 sets a circle centered on the center of the face as the reduced area 602. The correction area setting unit 42 sets the reduced area 602 so that the outer periphery of the reduced area 602 is near the contour of the human face 601. The correction area setting unit 42 sets the enlarged area 604 outside the reduced area 602. However, since the human face 601 is located at the right end, the correction area setting unit 42 cannot set the enlarged area 604 on the right side of the face 602.

FIG. 7B shows an image 605 obtained by correcting the face 601 of the person so that the face becomes smaller toward the center of the face with respect to the input image 600. The corrected face 606 of the person is corrected to be smaller than the face 601 of the person before correction. A region 607 indicated by shading at the right end of the image 605 indicates a region where the reference destination is outside the image of the input image 600, and the pixel value is indefinite. That is, the point P = (rcos θ, rsin θ) + (c1, c2) corresponding to P ′ = (r ′ cos θ, r ′ sin θ) + (c1, c2) in the region 607 is outside the input image 600. . The region 508 in which the pixel value is indefinite in the processing example 2 is a rectangle, whereas the region 607 in the processing example 4 is different in that the region is partially surrounded by a curve. This is due to the difference.

Therefore, the cut setting unit 44 sets not the area 607 but the rectangular area 603 including the area 607 as the cut area. As a result, as shown in FIG. 7C, it is possible to obtain a suitable corrected image that does not include a region with an indefinite pixel value and has a rectangular cut image.

<Processing example 5>
Next, the case where the correction target region is located at the corner of the input image will be described. FIG. 8A shows an input image 700 used in Processing Example 5, and a human face (target region) 701 is located at the upper right end. In the processing example 5, as in the processing example 4, the control unit 40 performs correction to reduce the human face 701 toward the center of the face. The correction area setting unit 42 sets a circle centered on the center of the face as the reduction area 702. The correction area setting unit 42 sets the reduced area 702 so that the outer periphery of the reduced area 702 is near the outline of the human face 701. The correction area setting unit 42 sets the enlarged area 704 outside the reduced area 702. However, since the human face 701 is located at the upper right end, the correction area setting unit 42 cannot set the enlarged area 704 on the right side of the face 702 and the upper side of the face 702.

8B shows an image 705 obtained by correcting the face 701 of the person so that the face becomes smaller toward the center of the face with respect to the input image 700. The corrected face 706 of the person is corrected to be smaller than the face 701 of the person before correction. In addition, two areas 707 indicated by shading at the right end and the upper end of the image 705 are areas where the reference destination is outside the image of the input image 700, and the pixel values are indefinite. That is, the point P = (rcos θ, rsin θ) + (c1, c2) corresponding to P ′ = (r ′ cos θ, r ′ sin θ) + (c1, c2) in the region 707 is outside the image of the input image 700. .

Therefore, the cut setting unit 44 sets two rectangular areas 703 ((b) and (d) in FIG. 8) including two areas 707 where pixel values are indefinite as cut areas. As a result, as shown in FIG. 8C, it is possible to obtain a suitable corrected image that does not include a region with an indefinite pixel value and that has a rectangular cut-out image.

As described above, in the present embodiment, when the correction region is located at the edge of the image and the reference destination of a part of the output image is outside the image of the input image, the pixel value is indefinite by referring to the outside of the image. By cutting out the output image so as to exclude the region to be, a suitable corrected image that does not include a region with an indefinite pixel value can be obtained.

In particular, since correction can be performed with a preset correction intensity regardless of the position of the correction area, the effect of the correction can be made constant regardless of the position of the correction area. Thus, when there are two human faces at different positions in the image, for example, at the center of the image and at the right end of the image, and the correction for reducing each face is performed, the same effect is obtained for the two persons. Obtained and preferred.

<Additional notes>
Note that another aspect of the present embodiment may be an image pickup apparatus that does not have a printing function. FIG. 9 is a functional block diagram illustrating a configuration of the imaging device 2 according to another aspect of the present embodiment. Similar to the image printing apparatus 1, the imaging apparatus 2 includes the imaging unit 10, the operation unit 20, the display unit 30, the control unit (image processing apparatus) 40, and the storage unit 60, but does not include the printing unit 50. . Further, the image processing apparatus may include only the control unit 40.

[Embodiment 2]
Another embodiment (Embodiment 2) of the present invention will be described below with reference to FIGS. For convenience of explanation, members having the same functions as those described in the embodiment are given the same reference numerals, and descriptions thereof are omitted.

In the second embodiment, the control unit 40 further cuts out an image after cutting out an area where the pixel value is indefinite, thereby generating a suitable output image having no change in the aspect ratio of the image before and after correction. Hereinafter, the details of the correction processing in the present embodiment will be described with reference to processing examples.

<Processing example 6>
FIG. 10A shows an image 800 after correction for reducing the face 801 is performed on an input image including a human face (target area) 801. A shaded area 802 is shown in FIG. An area where the pixel value is indefinite is shown. The cut setting unit 44 sets a rectangular area 805 including the area 802 as a cut area.

FIG. 10B shows an image 803 in which the region 805 has been cut out. The image 803 has a different aspect ratio from the image 800 because the right side of the image 800 is cut off. For example, if the number of pixels of the image 800 is 1600 pixels wide and 1200 pixels long, the aspect ratio of the image 800 is 4: 3. On the other hand, the image 803 has an aspect ratio different from that of the image 800 because the right side of the image 800 is cut off. If the number of pixels in the rectangular area cut out from the image 800 is 100 pixels wide and 1200 pixels high, the image 803 has 1500 pixels horizontally and 1200 pixels vertically, the aspect ratio is 5: 4, and the aspect ratio of the image 800 is Is different.

Therefore, in this embodiment, the cropping setting unit 44 adds the aspect ratio of the output image to the aspect ratio of the input image in addition to the rectangular area 805 including the area 802 where the pixel value is indefinite. As shown in (d) of FIG. 6, a cutout area 806 is set on the upper side of the image 800. As a result, the image 804 from which the cut area is cut by the image correction unit 45 is cut off by 1500 pixels in the horizontal direction and 75 pixels in the vertical direction as shown in FIG. The number of pixels is 1500 horizontal and 1125 vertical, and the aspect ratio is 4: 3. Therefore, the aspect ratio is the same as the aspect ratio of the image before correction.

The following effects can be obtained because the aspect ratio does not change. When an image is displayed on a display device such as a liquid crystal display, the entire image can be displayed if the aspect ratios of the display screens match. On the other hand, when the aspect ratio of the display screen does not match the aspect ratio of the image to be displayed, there is a region (margin region) where no image is displayed on a part of the display screen. Also, if the aspect ratio of each image is different due to image correction (cutout), displaying a plurality of images in sequence on the display screen has a problem that the blank area changes every time the image is switched. If the aspect ratios of a plurality of images match, it is preferable that the blank area does not change when the image to be displayed is switched. In addition to displaying on the display screen, when printing an image on paper with a printing device or printing as a photo, if the aspect ratio of the printing paper matches the aspect ratio of the image, It is preferable that printing can be performed without a blank area.

Note that the control unit 40 does not necessarily have to proceed with the correction in the order of (a) to (c) in FIG. 10, and the order may be changed or may be performed in parallel.

As described above, when the cropping setting unit 44 sets a cropping region on a certain side of the input image, the aspect ratio of the output image after correction is the same as the aspect ratio of the input image before correction. By cutting out the side (second side) that does not face the certain side (first side) in the input image, an output image maintaining the aspect ratio of the image can be obtained.

In one embodiment, the cropping setting unit 44 assumes that the number of horizontal pixels of the input image before correction is nx, the number of vertical pixels of the input image before correction is ny, and the width of the cropping area is w. If the area is close to the right or left side of the input image and the cut area is set on the right or left side, a cut area having a width of w (ny / nx) may be set on the upper or lower side. , (Ii) When the target area is close to the upper side or the lower side of the input image and the cut area is set on the upper side or the lower side, a cut area having a width of w (nx / ny) is further set on the right side or the left side. Should be set.

<Processing example 7>
Next, a method for setting a position to cut out an image in order to adjust the aspect ratio will be described. FIG. 11A shows an image 900 after correction for reducing the face 901 is performed on an input image including the face (target region) 901 of a person and the face 903 of another person. A shaded area 902 represents an area where the pixel value is indefinite. The cut setting unit 44 sets a rectangular area 905 including the area 902 as a cut area.

Here, in the present embodiment, the cropping setting unit 44 adds the aspect ratio of the output image to the aspect ratio of the input image in addition to the rectangular area 905 including the area 902 where the pixel value is indefinite. In 900, a cutting area is further set on the upper side or the lower side that is not opposite to the right side where the area 902 is set.

At this time, in the processing example 7, since the human face 903 is shown on the upper side of the image 900, the crop setting unit 44 has a region 906 on the lower side of the image 900 as shown in FIG. Is set as the cropping area. As a result, as shown in FIG. 11B, the image 904 from which the cut region is cut by the image correction unit 45 includes both the human face 901 and the human face 903, and the input image before correction. This is preferable because it has the same aspect ratio as the above aspect ratio.

Note that the criterion for selecting the cropping region from the cropping region candidate regions so that the cropping setting unit 44 does not change the aspect ratio of the image before and after correction is not limited to the criterion of whether or not a human face is included. . For example, the cut setting unit 44 determines whether or not a subject of interest other than a person is included in a candidate area of a cut area for which the aspect ratio is not changed before and after correction, A candidate area that does not include the subject of interest may be set as a cut-out area. In addition, the cutout setting unit 44 calculates the feature amount of each candidate region and cuts out the candidate region with a small feature amount so as to exclude the candidate face when neither of the candidate regions includes a human face or a target subject. Also good. The cut setting unit 44 can calculate the feature amount using color information such as edge detection and saturation, for example.

That is, the cropping setting unit 44 sets a cropping region having a specific width (first width) on the upper side or the lower side (or the right side or the left side) in order not to change the aspect ratio of the image before and after correction. A clipping region is set on the side of the upper side or the lower side (or the right side or the left side) where the target pixel is not included within a specific width from the side. Thereby, the cropping setting unit 44 can avoid losing important parts in the output image.

Here, the target pixel means a feature point for detecting a person, a target subject, or a pixel for detecting a feature amount.

<Processing Example 8>
Next, in order to adjust the aspect ratio, a method based on the position of a person's face may be used as a method for setting a position to cut out an image. FIG. 12A shows an image 1000 after correction for reducing the face 1001 is performed on an input image including a human face (target area) 1001, and an area 1002 indicated by shading is shown in FIG. An area where the pixel value is indefinite is shown. The cut setting unit 44 sets a rectangular area 1005 including the area 1002 as a cut area.

Here, in the present embodiment, the cropping setting unit 44 adds the aspect ratio of the output image to the aspect ratio of the input image in addition to the rectangular region 1005 including the region 1002 where the pixel value is indefinite. In 1000, a cutting area is further set on the upper side or the lower side that is not opposed to the right side where the area 1002 is set.

At this time, in the processing example 8, the crop setting unit 44 uses the region 1006 on the upper side of the image 1000, which is the side farther from the reduced region including the person 1001, as the crop region, as illustrated in FIG. Set. As a result, the image 1003 from which the crop region has been cut out by the image correction unit 45 is shown in FIG. 12B, in which the person's face 1001 is shown at a position close to the center of the image 1003. The target area to be formed can be made closer to the center of the image, and the same aspect ratio as the aspect ratio of the input image before correction is preferable.

As shown in FIG. 12E, when the crop setting unit 44 sets a region 1007 on the lower side of the image 1000, which is the side closer to the reduced region including the person 1001, as the crop region, the image correction unit. In the image 1004 in which the cut region is cut out by 45, the face 1001 of the person appears in a position away from the center of the image 1003 as shown in FIG.

[Embodiment 3]
Another embodiment (third embodiment) of the present invention will be described below with reference to FIG. For convenience of explanation, members having the same functions as those described in the embodiment are given the same reference numerals, and descriptions thereof are omitted.

In the third embodiment, the control unit 40 generates a more suitable output image by performing pixel number conversion on an image whose aspect ratio is adjusted. Hereinafter, the details of the correction processing in the present embodiment will be described with reference to processing examples.

<Processing example 9>
FIG. 13A shows an image 1100 used in the processing example 9 in which a human face (target region) 1101 is shown.

In FIG. 13B, as in the processing examples 6 to 8, image correction is performed to reduce the human face 1101, and the cropping area 1104 where the pixel value is indefinite and the aspect ratio are adjusted. An image 1102 obtained by cutting the cut area 1105 is shown. The image 1102 shown in FIG. 13B and the image 1100 shown in FIG. 13D have the same aspect ratio, but differ in the number of pixels of the image corresponding to the cut region. In that case, when an image is captured by the imaging apparatus, an image having a smaller number of pixels than the preset number of pixels of the image is output.

Therefore, in the present embodiment, the image correction unit 45 cuts out the cutout area 1104 in which the pixel value is indefinite and the cutout area 1105 to adjust the aspect ratio, whereby the number of vertical pixels and the horizontal length of the image are adjusted. When the number of pixels is different from the number of vertical pixels and the number of horizontal pixels of the input image, an image in which the cut region is cut out so that the number of vertical pixels and the number of horizontal pixels match between the input image and the output image 1102 is enlarged.

The image enlargement process can be executed by a known interpolation method such as the nearest neighbor method, the bilinear method, or the bicubic method. FIG. 13C shows an image 1103 enlarged from the image 1102 by the image correction unit 45. The image 1103 shown in FIG. 13C and the image 1100 shown in FIG. 13E have the same image size (number of pixels), and the human face 1101 is corrected to be small. Thereby, the image correction unit 45 can generate an image having the same number of pixels as the input image and in which the correction target subject is preferably corrected.

Note that the control unit 40 does not necessarily have to adjust the aspect ratio adjustment and the pixel number conversion by cropping in the order of (a) to (c) of FIG. You may do it.

[Embodiment 4]
The following will describe another embodiment (Embodiment 4) of the present invention with reference to FIG. For convenience of explanation, members having the same functions as those described in the embodiment are given the same reference numerals, and descriptions thereof are omitted.

The control unit 40 generates a suitable output image having no change in the aspect ratio of the image before and after correction by enlarging a part of the region after cutting out the region where the pixel value is indefinite. Hereinafter, the details of the correction processing in the present embodiment will be described with reference to processing examples.

<Process Example 10>
FIG. 14A is a diagram showing an image 1500 including a human face (target region) 1501 used in Processing Example 10. The correction area setting unit 42 sets a reduction area 1511 and an enlargement area 1512 for the image 1500.

FIG. 14B shows an image 1502 in which a correction for reducing the face 1501 is performed and a region 1505 with an indefinite pixel value is cut out. The image 1502 has a different aspect ratio from the image 1500 because the right side of the image 1500 is cut off.

Therefore, in the present embodiment, the correction area setting unit 42 applies an image to the reduced area 1511 as shown in FIG. 14B so that the aspect ratio of the output image is the same as the aspect ratio of the input image. A second enlarged region (third region) 1506 is set on the side (left side, third side) opposite to the side (right side, first side) from which is cut off. As a result, the image 1508 in which the width of the second enlarged region 1506 is enlarged from the width 1507 to the width 1510 by the image correction unit 45 is cut out as shown in FIG. To compensate for the decreased number of pixels in the horizontal direction, and the aspect ratio becomes the same as the aspect ratio of the image before correction.

Thus, unlike the second embodiment, the aspect ratio can be adjusted without performing unnecessary clipping. Note that the second enlarged region 1506 is preferably wide. If the second enlargement area 1506 is wide, the enlargement / reduction ratio β of the second enlargement area 1506 becomes a value close to 1, and the output image quality can be prevented from deteriorating.

In addition, the correction area setting unit 42 sets, as a second enlarged area 1506, an area that does not include the target pixel among areas located on the side opposite to the side where the image is cut out with respect to the reduced area. Is preferred. As a result, it is possible to avoid a human face or the like being included in the second enlarged region 1506 and deforming the face or the like.

[Embodiment 5]
The following will describe another embodiment (Embodiment 5) of the present invention with reference to FIGS. For convenience of explanation, members having the same functions as those described in the embodiment are given the same reference numerals, and descriptions thereof are omitted.

In the image processing apparatus according to the present embodiment, the reduced area approaches a certain side (first side) of the input image, and an enlarged area having a sufficient width cannot be set on the first side of the reduced area. By changing the reduction direction of the reduction area, it is possible to avoid the generation of an area with an indefinite pixel value. This is different from the case where the image processing apparatuses according to the first to fourth embodiments set the cut region on the first side.

FIG. 15 is a functional block diagram showing a configuration example of the control unit according to the present embodiment. The image printing apparatus according to the present embodiment has the same configuration as that of the image printing apparatus 1 according to the first embodiment, except that the control unit (image processing apparatus) 140 is provided instead of the control unit 40. As shown in FIG. 15, the control unit 140 includes a subject detection unit 141, a correction area setting unit 142, a correction direction setting unit 143, and an image correction unit 144. The correction area setting unit 142, the correction direction setting unit 143, and the image correction unit 144 are collectively referred to as a correction processing unit.

The subject detection unit 141 has the same function as the subject detection unit 41. The correction area setting unit 142 has the same function as the correction area setting unit 42. The correction direction setting unit 143 has the same function as that of the correction intensity setting unit 43, and further has a function of changing the reduction direction and the enlargement area of the reduction area and a function of setting the cut area, as will be described later. The image correction unit 144 has a function equivalent to that of the image correction unit 45.

Next, the flow of image processing in the control unit 140 of this embodiment will be described with reference to FIG. FIG. 16 is a flowchart illustrating an example of the flow of image processing in the control unit 140.

When the subject detection unit 141 acquires the image captured by the imaging unit 10 as an input image, the subject detection unit 141 detects a subject to be corrected from the input image (S201: target region detection step). The correction area setting unit 142 sets a reduction area and an enlargement area based on information indicated by the subject detected by the subject detection unit 41 (S202: correction processing step). Further, the correction direction setting unit 143 corrects the correction strength (the reduction magnification α in the reduction region and the enlargement magnification β in the enlargement region) based on the information indicating the reduction region and the enlargement region set by the correction region setting unit 42. Are set, and if necessary, some of the change of the reduction direction of the reduction region, the change of the enlargement region, and the setting of the cutout region are performed (S1603: correction processing step). Then, the image correction unit 144 generates an output image by reducing the reduction area, expanding the enlargement area, and cutting the cut area as necessary (S205: correction processing step). Then, the control unit 40 outputs the generated output image to the printing unit 50 or the display unit 30.

Subsequently, details of the correction processing by the control unit 140 will be described with reference to processing examples.

<Process Example 11>
FIG. 17A shows the input image 1700 used in the processing example 11 and the correction areas (the reduction area 1702 and the enlargement area 1703) set in the input image 1700 by the correction area setting unit 142. In the input image 1700, since the human face (target region) 1701 is located at the right end, the correction region setting unit 142 sets the reduced region 1702 so as to be in contact with the right side of the input image 1700. Therefore, the correction area setting unit 142 can set the enlarged area 1703 only on the left side of the reduced area 1702, and cannot set an enlarged area with a sufficient width on the right side of the reduced area 1702.

At this time, the correction direction setting unit 143 changes the reduction direction of the reduction area 1702 so as not to go from the right side of the input image 1700 to the inside of the reduction area 1702. For example, the correction direction setting unit 143 changes the reduction direction of the reduction area 1702 so as to be directed toward the side (the right side of the input image 1700) to which the reduction area 1702 is close. Further, the correction direction setting unit 143 resets the enlargement area so that the enlargement area is not set on the side close to the reduction area 1702 (the right side of the input image 1700).

FIG. 17B shows an output image 1704 obtained by correcting the input image 1700 so that the width of the face 1705 is reduced in the reduction direction changed by the correction direction setting unit 143. The width of the face 1705 after correction (= the width of the reduced area 1706 after correction) is corrected to be narrower than the width of the face 1701 before correction (= the width of the reduction area 1702 before correction). .

Here, since the reduction direction of the reduction region 1702 is rightward, a point P = (e−1, y) whose distance from the right end x = e of the reduction region 1702 is l is a point P ′ = (e− (αl, y) is corrected, and the right end of the reduced area 1706 after correction is also x = e. Therefore, it is not necessary to set an enlarged area on the right side of the reduced area 1702. Therefore, as in Reference Example 1, the reduced area and a certain side (first side) of the input image approach each other, and a sufficient enlarged area cannot be set on the first side of the reduced area. However, it is possible to avoid the occurrence of a region with an indefinite pixel value on the first side of the reduced region.

Note that if the left end of the pre-correction reduced area 1702 is x = ew, the left end of the post-correction reduced area 1706 is x = e−αw, which is shifted to the right side from x = ew, so that the image correction unit 144 shifts the enlarged area 1707 and the left area to the right. As a result, a region 1704 indicated by shading at the left end of the corrected output image 1704 is a region with an indefinite pixel value.

Therefore, the correction direction setting unit 143 sets the area 1704 as a cut-out area. As a result, as shown in FIG. 17C, the image correction unit 144 can generate an output image 1709 in which a region 1708 in which the pixel value is indefinite is cut out from the output image 1704. As a result, it is possible to obtain a suitable output image that does not include a region with an indefinite pixel value.

Note that the correction direction setting unit 143 also sets the pixel value by expanding the region sandwiched between the region 1704 and the enlarged region 1707 so as to supplement the region 1704 instead of setting the region 1704 as a cutout region. A suitable output image that does not include an indefinite area can be obtained.

As described above, the correction direction setting unit 143 according to the present embodiment, when the reduction direction of the reduction area is directed from one side (first side) side of the input image to the inside of the reduction area, the first direction from the reduction area. It is determined whether to change the reduction direction of the reduction area according to the distance to the side of the image, and when changing the reduction direction of the reduction area, the reduction direction of the reduction area is changed from the first side to the reduction area. Change so that it does not face inward. As a result, it is possible to obtain a suitable output image that does not include a region with an indefinite pixel value.

<Processing example 12>
FIG. 18A shows the input image 1800 used in the processing example 12 and the correction regions (the reduction region 1802 and the enlargement region 1803) set in the input image 1800 by the correction region setting unit 142. In the input image 1800, since the human face (target area) 1801 is located at the right end, the correction area setting unit 142 sets the reduced area 1802 so as to be in contact with the right side of the input image 1800. Therefore, the correction area setting unit 142 can set the enlarged area 1803 only on the left side of the reduced area 1802, and cannot set an enlarged area with a sufficient width on the right side of the reduced area 1802.

At this time, the correction direction setting unit 143 changes the reduction direction of the reduction area 1802 so as not to go to the inside of the reduction area 1802 from the right side of the input image 1800 as in the case of the processing example 11. The enlargement area is reset so that the enlargement area is not set on the side close to the reduction area 1802 (the right side of the input image 1800). Further, the correction direction setting unit 143 increases the enlargement ratio of the enlargement area 1803.

18B shows an output image 1804 obtained by correcting the input image 1800 so that the width of the face 1805 is reduced in the reduction direction changed by the correction direction setting unit 143. The width of the face 1805 after correction (= the width of the reduced area 1806 after correction) is corrected to be narrower than the width of the face 1801 before correction (= the width of the reduction area 1802 before correction). . The narrowed width of the reduced area 1802 is supplemented by the enlarged enlarged area 1803, and the area on the left side of the corrected enlarged area 1807 is not enlarged or reduced. As a result, it is possible to obtain a suitable output image that does not include a region with an indefinite pixel value.

[Example of software implementation]
Control blocks (particularly subject detection unit 41, correction region setting unit 42, correction strength setting unit 43, cutout setting unit 44, image correction unit 45, subject detection unit 141, correction region setting) of control units (image processing apparatuses) 40 and 140 Unit 142, correction direction setting unit 143, and image correction unit 144) are logic circuits (hardware) formed in an integrated circuit (IC chip) such as an application specific integrated circuit (ASIC) or a field programmable gate array (FPGA). Alternatively, it may be realized by software, or may be realized by software using a CPU (Central Processing Unit) or a GPU (Graphics Processing Unit).

In the latter case, the control units (image processing apparatuses) 40 and 140 have a CPU that executes instructions of a program that is software that realizes each function, and the program and various data are recorded so as to be readable by a computer (or CPU). A ROM (Read Only Memory) or a storage device (these are referred to as “recording media”), a RAM (Random Access Memory) for expanding the program, and the like are provided. And the objective of this invention is achieved when a computer (or CPU) reads the said program from the said recording medium and runs it. As the recording medium, a “non-temporary tangible medium” such as a tape, a disk, a card, a semiconductor memory, a programmable logic circuit, or the like can be used. The program may be supplied to the computer via an arbitrary transmission medium (such as a communication network or a broadcast wave) that can transmit the program. Note that one embodiment of the present invention can also be realized in the form of a data signal embedded in a carrier wave, in which the program is embodied by electronic transmission.

[Summary]
The image processing apparatus (control unit 40) according to aspect 1 of the present invention reduces the first area (reduction area) in the image and enlarges the second area (enlargement area) outside the first area, A correction processing unit (correction region setting unit 42, correction intensity setting unit 43, crop setting unit 44, image correction unit 45) for correcting the image is provided, and the correction processing unit is configured such that the reduction direction of the first region is Whether to cut out the first side of the image according to the distance from the first region to the first side when going from the first side of the image to the inside of the first region decide.

In the above configuration, the reduction direction of the first region (reduction region) is from the first side of the image to the inside of the first region, and the distance from the first region to the first side is short, When the second region (enlarged region) having a sufficient width cannot be provided on the first side of one region, a region with an indefinite pixel value may occur on the first side of the output image. Therefore, depending on the distance from the first region to the first side, a suitable output image that does not include a region with an indefinite pixel value can be obtained by cutting out the first side of the image. .

The image processing apparatus according to aspect 2 of the present invention is the image processing apparatus according to aspect 1, wherein the correction processing unit detects the first side of the image when the first region is in contact with the first side. It may be cut out.

According to the above configuration, the first area (reduced area) is in contact with the first side of the input image, and the second area (enlarged area) is completely located on the first side of the first area (reduced area). Even if it is not possible to set, a suitable output image that does not include a region with an indefinite pixel value can be obtained.

The image processing device according to aspect 3 of the present invention is the image processing apparatus according to aspect 1 or 2, wherein the correction processing unit is configured such that the distance from the first region to the first side is the first region across the first region. When the width of the second region on the opposite side to the side is narrower, the first side of the image may be cut out.

According to the above configuration, the first region (reduced region) is close to the first side of the input image, and the second region (enlarged region) having a sufficient width on the first side of the first region (reduced region). ) Cannot be set, it is possible to obtain a suitable output image that does not include a region with an indefinite pixel value.

The image processing apparatus according to aspect 4 of the present invention is the image processing apparatus according to aspects 1 to 3, wherein the correction processing unit is configured so that the aspect ratio of the image after correction is the same as the aspect ratio of the image before correction. A second side that is not opposed to the first side in the image may be cut out.

According to the above configuration, an output image maintaining the image aspect ratio can be obtained.

The image processing apparatus according to aspect 5 of the present invention is the image processing apparatus according to aspect 4, wherein the correction processing unit cuts the second side with a first width, and does not face the first side of the image. Of these, a side from which the target pixel is not included within the first width may be the second side.

According to the above configuration, it is possible to avoid losing important parts in the output image.

The image processing apparatus according to aspect 6 of the present invention is the image processing apparatus according to aspect 4, wherein the correction processing unit selects a side farther from the first region among the sides that do not face the first side in the image. The second side may be used.

According to the above configuration, in the output image, it is possible to bring the target to be noticed closer to the center of the image.

In the image processing device according to aspect 7 of the present invention, in any one of the above aspects 4 to 6, the correction processing unit corrects the number of pixels in the vertical direction and the number of pixels in the horizontal direction of the image after correction. The entire image obtained by cutting out the first side and the second side may be enlarged so that the number of pixels in the vertical direction and the number of pixels in the horizontal direction of the previous image are the same. .

According to the above configuration, it is possible to obtain an output image that maintains the number of pixels in the vertical direction and the number of pixels in the horizontal direction of the image.

In the image processing apparatus according to aspect 8 of the present invention, in the above aspects 1 to 3, the correction processing unit is configured so that the aspect ratio of the image after correction is the same as the aspect ratio of the image before correction. The third region located on the third side facing the first side with respect to the first region may be enlarged.

According to the above configuration, an output image maintaining the image aspect ratio can be obtained.

The image processing device according to aspect 9 of the present invention is the image processing apparatus according to aspect 8, wherein the correction processing unit includes an area that does not include a target pixel in an area located on the third side with respect to the first area. The third region may be used.

According to the above configuration, it is possible to obtain a high-quality output image that maintains the aspect ratio without deforming important parts.

The image processing apparatus (control unit 140) according to the aspect 10 of the present invention reduces the first area (reduction area) in the image and enlarges the second area (enlargement area) outside the first area. A correction processing unit (a correction region setting unit 142, a correction direction setting unit 143, and an image correction unit 144) that corrects the image is provided, and the correction processing unit is configured such that the reduction direction of the first region is the first of the image. When going from the side to the inside of the first region, the reduction direction of the first region is changed from the first side of the image to the first side according to the distance from the first region to the first side. It is determined whether or not to change so as not to go inside one area.

In the above configuration, the reduction direction of the first region (reduction region) is from the first side of the image to the inside of the first region, and the distance from the first region to the first side is short, When the second region (enlarged region) having a sufficient width cannot be provided on the first side of one region, a region with an indefinite pixel value may occur on the first side of the output image. Therefore, depending on the distance from the first region to the first side, the pixel value is changed by changing the reduction direction of the first region so as not to go from the first side of the image to the inside of the first region. A suitable output image that does not include an indefinite area can be obtained.

An imaging apparatus 2 according to an aspect 11 of the present invention includes an imaging unit 10 and an image processing apparatus according to any one of the above aspects 1 to 10 that corrects the image captured by the imaging unit 10. Yes.

According to the above configuration, the user can capture an image of a person's face and suitably perform image processing on the captured image.

An image printing apparatus 1 according to an aspect 12 of the present invention includes the image processing apparatus according to any one of the above aspects 1 to 10 and a printing unit 50 that prints the image corrected by the image processing apparatus. Yes.

According to the above configuration, the user can easily print an image subjected to image processing.

The image printing apparatus 1 according to the aspect 13 of the present invention is an image processing apparatus according to any one of the above aspects 1 to 10, which performs image processing on the imaging unit 10 and the image captured by the imaging unit 10. And a printing unit 50 that prints the image corrected by the image processing apparatus.

According to the above configuration, the user can easily print an image obtained by performing image processing on the captured image.

A control method of an image processing device according to aspect 14 of the present invention includes an image correction step of correcting the image by reducing a first region in the image and enlarging a second region outside the first region. In the image correction step, when the reduction direction of the first region is directed from the first side to the inside of the first region, according to the distance from the first region to the first side, It is determined whether or not to cut out the first side in the image.

In addition, in the control method of the image processing device according to the aspect 15 of the present invention, the image correction step of correcting the image by reducing the first region in the image and expanding the second region outside the first region. In the image correction step, when the reduction direction of the first region is directed from the first side to the inside of the first region, the image correction step depends on a distance from the first region to the first side. Then, it is determined whether or not to change the reduction direction of the first region so as not to go from the first side of the image to the inside of the first region.

According to the above configuration, the same effects as those of the image processing apparatus according to one aspect of the present invention can be obtained.

Furthermore, the image processing apparatus according to each aspect of the present invention may be realized by a computer. In this case, the image processing apparatus is operated by causing the computer to operate as each unit (software element) included in the image processing apparatus. An image processing program for an image processing apparatus realized by a computer and a computer-readable recording medium on which the image processing program is recorded also fall within the scope of the present invention.

[Additional Notes]
The present invention is not limited to the above-described embodiments, and various modifications are possible within the scope shown in the claims, and embodiments obtained by appropriately combining technical means disclosed in different embodiments. Is also included in the technical scope of the present invention. Furthermore, a new technical feature can be formed by combining the technical means disclosed in each embodiment.

(Cross-reference of related applications)
This application claims the benefit of priority to the Japanese patent application filed on Apr. 20, 2017: Japanese Patent Application No. 2017-084003. Included in this document.

DESCRIPTION OF SYMBOLS 1 Image printing apparatus 2 Imaging device 10 Imaging part 40,140 Control part (image processing apparatus)
41, 141 Subject detection unit (target region detection unit)
42, 142 Correction area setting section (correction processing section)
43 Correction strength setting section (correction processing section)
44 Cutting setting section (correction processing section)
45, 144 Image correction unit (correction processing unit)
143 Correction direction setting unit (correction processing unit)
50 printing section

Claims (16)

1. A correction processing unit that corrects the image by reducing the first region in the image and enlarging the second region outside the first region;
   When the reduction direction of the first region is from the first side of the image toward the inside of the first region, the correction processing unit is responsive to a distance from the first region to the first side. An image processing apparatus that determines whether to cut out the first side of the image.
2. The image processing apparatus according to claim 1, wherein the correction processing unit cuts out the first side of the image when the first region and the first side are in contact with each other.
3. When the distance from the first region to the first side is narrower than the width of the second region opposite to the first side across the first region, the correction processing unit, The image processing apparatus according to claim 1, wherein the first side of the image is cut out.
4. The correction processing unit cuts a second side of the image that does not face the first side so that the aspect ratio of the image after correction is the same as the aspect ratio of the image before correction. The image processing apparatus according to any one of claims 1 to 3, wherein:
5. The correction processing unit cuts out the second side with a first width, and among the sides that do not face the first side in the image, the pixel of interest is included in the first width from the side. The image processing apparatus according to claim 4, wherein a side that does not exist is the second side.
6. 5. The correction processing unit according to claim 4, wherein among the sides that do not face the first side in the image, the side farther from the first region is set as the second side. 6. Image processing device.
7. The correction processing unit is configured so that the number of vertical pixels and the number of horizontal pixels of the image after correction are the same as the number of vertical pixels and the number of horizontal pixels of the image before correction. The image processing apparatus according to any one of claims 4 to 6, wherein the entire image obtained by cutting out the first side and the second side is enlarged.
8. The correction processing unit has a third side facing the first side with respect to the first region so that the aspect ratio of the image after correction is the same as the aspect ratio of the image before correction. The image processing apparatus according to any one of claims 1 to 3, wherein the third region located on the side is enlarged.
9. The said correction process part makes the area | region which does not contain a pixel of interest among the area | regions located in the said 3rd edge | side side with respect to the said 1st area | region as said 3rd area | region. Image processing device.
10. A correction processing unit that corrects the image by reducing the first region in the image and enlarging the second region outside the first region;
    When the reduction direction of the first region is from the first side of the image toward the inside of the first region, the correction processing unit is responsive to a distance from the first region to the first side. An image processing apparatus that determines whether or not to change the reduction direction of the first area so as not to go from the first side of the image to the inside of the first area.
11. An imaging unit;
    An image processing apparatus comprising: the image processing apparatus according to claim 1 that corrects the image captured by the image capturing unit.
12. The image processing apparatus according to any one of claims 1 to 10,
    An image printing apparatus comprising: a printing unit that prints the image corrected by the image processing apparatus.
13. An imaging unit;
    The image processing apparatus according to any one of claims 1 to 10, wherein image processing is performed on the image captured by the imaging unit.
    An image printing apparatus comprising: a printing unit that prints the image corrected by the image processing apparatus.
14. Including an image correction step of correcting the image by reducing a first region in the image and enlarging a second region outside the first region;
    In the image correction step, when the reduction direction of the first region is directed from the first side to the inside of the first region, according to the distance from the first region to the first side, A method for controlling an image processing apparatus, comprising: determining whether or not to cut out the first side in an image.
15. Including an image correction step of correcting the image by reducing a first region in the image and enlarging a second region outside the first region;
    In the image correction step, when the reduction direction of the first region is directed from the first side to the inside of the first region, according to the distance from the first region to the first side, A method for controlling an image processing apparatus, comprising: determining whether or not to change the reduction direction of the first area so as not to go from the first side of the image toward the inside of the first area.
16. An image processing program for causing a computer to function as the image processing apparatus according to any one of claims 1 to 10, wherein the image processing program causes the computer to function as the correction processing unit.

Images